If a nonblocking write requires short writes, just try it once
as in the blocking case. After all, we are nonblocking, so
don't loop unnecessarily.
Fixes: 170e6badb621 ("Cygwin: pipe: improve writing when pipe buffer is almost full")
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
There's no reason to overwrite len1 with align if they are the same value.
Fixes: 170e6badb621 ("Cygwin: pipe: improve writing when pipe buffer is almost full")
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
So far fhandler_pipe_fifo::raw_write always returns -1/EINTR
if a signal arrived. Linux does not do that if there's still
space left in the pipe buffer.
The Linux buffer handling can't be emulated by Cygwin, but we
can do something similar which makes it much more likely to still
write successfully even if the buffer is almost full.
Utilize pipe_data_available to return valid pipe buffer usage
to raw_write, allowing a more sophisticated way to fill the
buffer while maintaining comaptibility with non-Cygwin pipes.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Linux select(2) returns the pipe as writable if at least one
free page (4K onl most systems) is left in a page-oriented buffer
handling. This is the same as PIPE_BUF.
Emulate this behaviour by only returning the pipe as writable
if at least 4K space is left in the buffer.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
The simple cygwait call in fhandler_pipe_fifo::raw_write doesn't
take the SA_RESTART setting into account. Move handling the
signal into raw_write.
Fixes: 4b25687ea3ee2 ("Cygwin: fhandler_pipe: add raw_read and raw_write")
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Rewrite to avoid new use-after-free warnings about realloc(), seen with
gcc 12, e.g.:
> In function ‘void rebase_locale_buf(const void*, const void*, const char*, const char*, const char*)’,
> inlined from ‘int __set_lc_time_from_win(const char*, const lc_time_T*, lc_time_T*, char**, wctomb_p, const char*)’ at ../../../../src/winsup/cygwin/nlsfuncs.cc:705:25:
> ../../../../src/winsup/cygwin/nlsfuncs.cc:338:24: error: pointer ‘new_lc_time_buf’ may be used after ‘void* realloc(void*, size_t)’ [-Werror=use-after-free]
> 338 | *ptrs += newbase - oldbase;
> | ~~~~~~~~^~~~~~~~~
> ../../../../src/winsup/cygwin/nlsfuncs.cc: In function ‘int __set_lc_time_from_win(const char*, const lc_time_T*, lc_time_T*, char**, wctomb_p, const char*)’:
> ../../../../src/winsup/cygwin/nlsfuncs.cc:699:44: note: call to ‘void* realloc(void*, size_t)’ here
> 699 | char *tmp = (char *) realloc (new_lc_time_buf, len);
Technically, it's UB to later refer to the realloced pointer (even just
for offset computations, without deferencing it), so switch to using
malloc() to create the resized copy.
Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
Fix gcc 12 warnings about narrowing conversions of socket ioctl constants
when used as case labels, e.g:
> ../../../../src/winsup/cygwin/net.cc: In function ‘int get_ifconf(ifconf*, int)’:
> ../../../../src/winsup/cygwin/net.cc:1940:18: error: narrowing conversion of ‘2152756069’ from ‘long int’ to ‘int’ [-Wnarrowing]
Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
The _IOR anbd _IOW macros cast the sizeof of their third arg
uselessly to long. The _IO definitions have been taken from BSD
in 1995 and never changed again. The fact that the sizeof() gets
cast to long is probably a remnant from the past when the stuff
was supposed to be used on 16 bit machines, but the value was
supposed to be 32 bit.
Given that the values are not supposed to be ever bigger than 32 bit,
we drop the (long) cast. Compare with current FreeBSD, which does
not cast at all.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Fix a gcc 12 warning about a narrowing conversion in case labels for
tape options.
> In file included from /wip/cygwin/src/winsup/cygwin/include/sys/mtio.h:14,
> from ../../../../src/winsup/cygwin/fhandler/tape.cc:13:
> ../../../../src/winsup/cygwin/fhandler/tape.cc: In member function ‘int mtinfo_drive::set_options(HANDLE, int32_t)’:
> ../../../../src/winsup/cygwin/fhandler/tape.cc:965:12: error: narrowing conversion of ‘4026531840’ from ‘unsigned int’ to ‘int’ [-Wnarrowing]
Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
Fix a gcc 12 warning about an address known to be non-NULL in
format_proc_locale_proc().
> ../../../../src/winsup/cygwin/fhandler/proc.cc: In function ‘BOOL format_proc_locale_proc(LPWSTR, DWORD, LPARAM)’:
> ../../../../src/winsup/cygwin/fhandler/proc.cc:2156:11: error: the address of ‘iso15924’ will never be NULL [-Werror=address]
> 2156 | if (iso15924)
> | ^~~~~~~~
> ../../../../src/winsup/cygwin/fhandler/proc.cc:2133:11: note: ‘iso15924’ declared here
> 2133 | wchar_t iso15924[ENCODING_LEN + 1] = { 0 };
> | ^~~~~~~~
Fixes: c42b98bdc665f ("Cygwin: introduce /proc/codesets and /proc/locales")
Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
Fix warnings with gcc 12 about narrowing conversions of NTSTATUS
constants when used as case labels, e.g:
> ../../../../src/winsup/cygwin/exceptions.cc: In static member function ‘static int exception::handle(EXCEPTION_RECORD*, void*, CONTEXT*, PDISPATCHER_CONTEXT)’:
> ../../../../src/winsup/cygwin/exceptions.cc:670:10: error: narrowing conversion of ‘-1073741682’ from ‘NTSTATUS’ {aka ‘int’} to ‘unsigned int’ [-Wnarrowing]
See also: c5bdf60ac46401a51a7e974333d9622966e22d67
Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
This disables a warning seen with gcc 12 caused by intrinsics used by
the inline implementation of NtCurrentTeb() inside w32api headers.
> In function ‘long long unsigned int __readgsqword(unsigned int)’,
> inlined from ‘_TEB* NtCurrentTeb()’ at /usr/include/w32api/winnt.h:10020:86,
> [...]
> /usr/include/w32api/psdk_inc/intrin-impl.h:838:1: error: array subscript 0 is outside array bounds of ‘long long unsigned int [0]’ [-Werror=array-bounds]
See also: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105523#c6
Signed-off-by: Jon Turney <jon.turney@dronecode.org.uk>
Linux 6.10 changed the content of cpufeatures.h to require explicit
quoted flag names for output in comments, instead of requiring a null
quoted string "" at the start of comments to suppress flag name output.
As a result, some flags (not all for output) were renamed and others moved:
- change dts to ds; move intel_ppin down; swap ibpd and ibrs;
- change some flag names and descriptions that are not output.
Signed-off-by: Brian Inglis <Brian.Inglis@SystematicSW.ab.ca>
Linux cpuinfo follows output for each processor with a blank line,
so we output newlines to get a blank line:
- newline after power management feature flags if printed;
- newline to give blank line after each processor output.
Reported-by: Achim Gratz https://cygwin.com/pipermail/cygwin/2024-July/256223.html
Signed-off-by: Brian Inglis <Brian.Inglis@SystematicSW.ab.ca>
If the master process of GNU screen or tmux is started in ConEmu
and ConEmu is closed, reattaching to the GNU screen/tmux in another
console will not work correctly. This is because the ConEmu master
process was already closed even though some console APIs are hooked
by ConEmuHk64.dll to communicate with ConEmu master process. With
this patch, to make them unhooked, DllMain() of ConEmuHk64.dll is
called with DLL_PROCESS_DETACH.
Fixes: 3721a756b0d8 ("Cygwin: console: Make the console accessible from other terminals.")
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
Previously, if two or more processes request to allocate a console at
the same time, the same device number could be accidentally allocated.
This patch fixes the issue by creating a named mutex (input_mutex) and
checking if GetLastError() is ERROR_ALREADY_EXIST or not to ensure
the unit number will be allocated in an atomic way. Thanks to this,
EnumWindow() is not necessary anymore to scan console units.
This also makes minor device numbers unique and console devices visible
across sessions. This disallows duplicated device number for different
sessions, so the MAX_CONS_DEV has been increased from 64 to 128.
Additionally, the console allocation and scanning will be faster as
a side effect of eliminating EnumWindows().
Fixes: 3721a756b0d8 ("Cygwin: console: Make the console accessible from other terminals.")
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
The commit baf2764dca43 has a serious bug that shared_info_state is
common for all console devices despite it should be individual for
each console device. This patch fixes that.
Fixes: baf2764dca43 ("Cygwin: console: Fixes an issue that tmux can not run on the console.")
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
After the commit c77a5689f7bd, tmux can not run on the console.
This patch replaces the countermeasure for the race issue between
console setup and close with another mechanism using a mutex.
Fixes: c77a5689f7bd ("Cygwin: console: Do not unmap shared console memory belonging to ctty.")
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
Previously, shared names in the console were created using get_minor().
However, get_minor() was not unique to the console across sessions.
This is because EnumWindows(), which is used to look for console windows,
cannot enumerate windows across sessions. This causes conflict on the
shared names between sessions (e.g. sessions of different users,
different services, a service and a user session, etc.).
With this patch, GetConsoleWindow() is used instead of get_minor().
GetConsoleWindow() has been used for the name of shared memory, which
should be unique to each console.
Addresses: https://cygwin.com/pipermail/cygwin/2024-April/255893.html
Fixes: ff4440fcf768 ("Cygwin: console: Introduce new thread which handles input signal.");
Reported-by: Johannes Khoshnazar-Thoma <johannes@johannesthoma.com>
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
Previsouly, the following commands hangs:
mintty -e timeout 1 dash -c 'yes aaaaaaaaaaaaaaaaaaaaaaaaa | cat'
The mechanism is as follows.
When the child process (timeout) is terminated, mintty seems to stop
reading pty master even if yes or cat still alive.
If the the pipe used to transfer output from pty slave to pty master
is full due to lack of master reader, WriteFile() to the pipe is
blocked. WriteFile() cannot be canceled by cygwin signal, therefore,
pty slave hangs.
This patch avoids hanging by checking pipe space before calling
WriteFile() and prevents writing data more than space.
Addresses: https://cygwin.com/pipermail/cygwin/2024-June/256178.html
Reported-by: jojelino <jojelino@gmail.com>
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
Previously, the following steps failed with error:
1) Open /dev/dsp with O_RDONLY
2) Open /dev/dsp with O_WRONLY
3) Issue SNDCTL_DSP_GETOSPACE ioctl() for 2)
This is because IS_WRITE() returns false for 2) due to incorrect
openflags handling in archetype instance. This patch fixes the
issue by adding open_setup() to fhandler_dev_dsp to set openflags
correctly for each instance.
Fixes: 92ddb7429065 ("* fhandler_dsp.cc (fhandler_dev_dsp::open): Remove archetype handling.")
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
w32api 12.0.0 adds the returns_twice attribute to RtlCaptureContext().
There's some data-flow interaction with using it inside a while loop
which causes a maybe-uninitialized warning.
../../../../winsup/cygwin/exceptions.cc: In member function 'int _cygtls::call_signal_handler()':
../../../../winsup/cygwin/exceptions.cc:1720:33: error: '<anonymous>' may be used uninitialized in this function [-Werror=maybe-uninitialized]
If ldd is run against a DLL which links to the Cygwin DLL, ldh will end
up loading the Cygwin DLL dynamically, much like cygcheck or strace.
Addresses: https://cygwin.com/pipermail/cygwin/2024-May/255991.html
Fixes: 60675f1a7eb2 ("Cygwin: decouple shared mem regions from Cygwin DLL")
Reviewed-by: Ken Brown <kbrown@cornell.edu>, Takashi Yano <takashi.yano@nifty.ne.jp>
Signed-off-by: Jeremy Drake <cygwin@jdrake.com>
To avoid race issues, pthread::once() uses pthread_mutex. This caused
the handle leak which was fixed by the commit 2c5433e5da82. However,
this fix introduced another race issue, i.e., the mutex may be used
after it is destroyed. This patch fixes the issue. Special thanks to
Bruno Haible for discussing how to fix this.
Addresses: https://cygwin.com/pipermail/cygwin/2024-May/255987.html
Reported-by: Bruno Haible <bruno@clisp.org>
Fixes: 2c5433e5da82 ("Cygwin: pthread: Fix handle leak in pthread_once.")
Reviewed-by: Ken Brown <kbrown@cornell.edu>
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
Currently, if the device does not have capability WAVECAPS_LRVOLUME,
the volume control does not work properly. This patch fixes that.
Fixes: 2a4af3661470 ("Cygwin: Implement sound mixer device.")
Signed-off-by: Takashi Yano <takashi.yano@nifty.ne.jp>
While commit 0321fb573654 ("Cygwin: glob: convert wchar_t to wint_t")
switched the entire glob mechanism from wchar_t to wint_t, the globify
function calling glob on a DOS command line did not catch up and only
checked for the next character constituting a 16 bit wchar_t. This
broke reading surrogate pairs from incoming DOS command lines if the
string had to go through glob(3).
Check for an entire 32 bit unicode char instead.
Fixes: 0321fb573654 ("Cygwin: glob: convert wchar_t to wint_t")
Reported-by: David Allsopp <david@tarides.com>
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Commit a0a25849f9dd ("Cygwin: fhandler_virtual: move fileid to path_conv
member") broke `ls -l /proc/<PID>'. Turns out, the commit forgot to set
the fileid in case of a virtual root dir (i. e., returning virt_rootdir)
in fhandlers utilizing fileid. This crashed opendir() due to a random
fileid.
Make sure to set fileid in any case.
Fixes: a0a25849f9dd ("Cygwin: fhandler_virtual: move fileid to path_conv member")
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Since we're opening the parent dir, it doesn't make sense
to print the name of the file in debug output. Print parent
dirname instead.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
If FILE_DIRECTORY_FILE is given, FILE_OPEN_NO_RECALL is not allowed,
otherwise NtCreateFile returns STATUS_INVALID_PARAMETER.
Drop FILE_OPEN_NO_RECALL where FILE_DIRECTORY_FILE is specified.
Fixes: f6b56abec186 ("Cygwin: try to avoid recalling offline files")
Reported-by: Bruce Jerrick <bmj001@gmail.com>
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Commit 80f722e97cf7 ("Cygwin: opendir(3): move ENOTDIR check into main
function") introduced a bug in fhandler_virtual handling. While the
assertion that path_conv::check() already calls exists() and sets
FILE_ATTRIBUTE_DIRECTORY accordingly, the exists() function is called
on a fhandler_virtual object created for just this code snippet. The
side effect of this is that the fileid member in the calling
fhandler_virtual object is not set after path_conv::check().
Move the fhandler_virtual::fileid member to path_conv::_virt_fileid
and create matching path_conv::virt_fileid() and fhandler_virtual::fileid()
methods.
Let path_conv::check() propagate the fileid set in the local
fhandler_virtual::exists() call to its own _virt_fileid.
Use new fhandler_virtual::fileid() method throughout.
Fixes: 80f722e97cf7 ("Cygwin: opendir(3): move ENOTDIR check into main function")
Reported-by: Bruce Jerrick <bmj001@gmail.com>
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Simplify code in that it only fetches a single entry per
IEnumShellItems::Next call. For some reason this appears to
be quicker most of the time.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Root dirs of WSL installations are exposed via \\wsl$.
Add support to expose \\wsl$.
While at it, check if TERMSRV and P9 provider exist before
exposing them.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
The //tsclient provider is WNNC_NET_TERMSRV.
Fixes: 7db1c6fc4e2a ("Cygwin: //server: revert to using WNet and support NFS shares")
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Continue using WSD for enumerating //, but switch back to
WNet for enumerating shares.
Add trying to enumerate NFS shares for machine names given as FQDN.
Only downcase server names, keep case of shares intact.
Move downcasing into dir_cache class.
Add a comment to explain an extrem weirdness in Windows.
Fixes: 205190a80bd0 ("Cygwin: // and //server: switch to Network Discovery")
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Checking server existence by trying to enumerate its shares
may result in 2 minutes delay until some internal timeout is hit.
In the light that every network is an IP network anyway these
days, let's try with a simple getaddrinfo() call. This is usually
back in 3 secs even if the server doesn't exist, and it's usually
back in 8 secs if the DNS server can't be connected. This is the
fastest method I found to check server existence yet.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
E. Europe Standard Time is available in the unicode.org windowsZones.xml
file, so we can drop the default handling.
Fix a few comments.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
Use official GetDynamicTimeZoneInformation() function instead of
scanning the registry for the timezone keyname.
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
The previous change introduced a missing end of comment
Fixes: 60ea9c1c4b66 ("Cygwin: tzmap-from-unicode.org: drop support for Vista/2008")
Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
